An evaluation of Intensive Interaction in community living
settings for adults with profound intellectual disabilities
Judith C. Samuel1 & Melanie A.E. Nind2*
with Amy Volans3 & Issy Scriven4
1
Judith Samuel, DClinPsychol, Consultant Clinical Pyschologist, Ridgeway Partnership
Oxfordshire Learning Disability NHS Trust, Slade House, Horspath Driftway,
Headington Oxford, OX3 7JH, UK.
Tel 01865 228066 Judith.Samuel@oldt.nhs.uk
2*
Melanie Nind, PhD, Deputy Head of School, University of Southampton, Highfield,
Southampton SO17 1BJ, UK
Tel 023 8059 5813 Fax 023 8059 3556 M.A.Nind@soton.ac.uk
3,4
Maudsley Hospital, UK
*Author for reprints and correspondence
Keywords: community living, intellectual disabilities, Intensive Interaction
1
An evaluation of Intensive Interaction in community living settings for
adults with profound intellectual disabilities
Abstract
Intensive Interaction is an approach to enhancing the communication and social abilities
of people with profound intellectual disabilities using principles from caregiver-infant
interaction. Use of Intensive Interaction by novice practitioners with four women with
profound intellectual disabilities living in a supported housing service in England was
evaluated. It was hypothesised that staff could learn the principles of Intensive Interaction
sufficiently to have a positive impact on the participants’ abilities (during five sessions
per week over twenty weeks) and that they would experience a better relationship. A
quasi-experimental interrupted time-series multiple-baseline design was used with data
collected via video observations, assessment schedules and staff questionnaires.
Practitioners learned to use mirroring and contingent responding and participants’
abilities to look at faces and become engaged in interaction and joint focus developed.
There was less volume of evidence for improved quality of relationship. Embedding
practitioner reflection on interactions was a difficult part of the intervention to
implement.
2
Introduction
The intervention
Intensive Interaction (Nind & Hewett, 1988; 2001; 2005) is a transactional approach
aimed at enhancing the responsiveness of teachers, carers and other communication
partners and thereby improving the communication and social abilities of people with
severe, complex or profound intellectual disabilities. It applies a (Western) model of
optimum caregiver-infant interaction to people with intellectual disabilities of any age.
Nind and Hewett developed Intensive Interaction based on Ephraim’s (1986) ‘augmented
mothering’. Features borrowed from caregiver-infant interaction include fostering mutual
pleasure and interactive games; altering interpersonal behaviour to become more
engaging and meaningful; making micro-adjustments to the flow of interactions with
pauses, repetitions and blended rhythms; imputing intentionality to pre-communicative
behaviours and using contingent responding (Nind, 1996). Attention is paid to the
professional elements of planning, monitoring and reflection, making it differ from the
intuitive approach in parenting.
Evidence
Evidence for the efficacy of Intensive Interaction is accumulating. First, using a multiplebaseline across participants interrupted time-series Nind (1996) evaluated Intensive
Interaction with three men and three women, aged 27-36 years, who were
institutionalised. The practitioners were teachers who were very familiar with the
approach and the participants conducted twice daily sessions for 12-18 months.
3
Participants showed new behaviours that encouraged others to be with them and a trend
towards a reduction in stereotypy; time spent in interactive social behaviour also
increased.
Lovell et al.'s (1998) short-term single-case experiment with a 53-year old man in a longstay hospital where Intensive Interaction was unfamiliar involved an expert practitioner
not known to the man. The alternating treatment design (five-minute Intensive Interaction
sessions and five minutes of ‘proximity’) found increased frequency of initiating physical
contact, looking at people, joint attention, smiling/laughing and vocalising in both
conditions but greater for Intensive Interaction. Initial success was not subsequently
transferred to novice staff. Elgie and Maguire's (2001) single-case experiment with a 39year old blind woman with self-injurious behaviour in a small residential home showed
an increase in spontaneous hand contact and vocalisation during sessions but no change
in self-injurious behaviour. Leaning and Watson's (2006) eight-week study of five people
found that a number of interactive behaviours increased from baseline during the
Intensive Interaction group and then reduced at follow up.
Research on Intensive Interaction with children includes Watson's (1994; Watson &
Knight, 1991) study of six staff-pupil pairs aged 10-19 throughout a school year which
found enhanced development and Kellett's (2000, 2001, 2003; Kellett & Nind, 2001,
2003) replication of Nind (1996). Kellett studied six children aged 4-11 years interacting
with novice practitioners - three teachers and three learning support assistants (all
women) who knew the participants. All participants developed their communication and
social abilities and the study illuminated features of service context required for effective
4
practice including management support, team approach, co-ordination, and intervention
continuity and fidelity.
Each effectiveness study is context-bound and the heterogeneity of people with profound
intellectual disabilities weakens any claim to generalisability. Studies vary in the
robustness of their design, participant and setting characteristics, practitioner expertise
and intervention duration, frequency and content. Nevertheless they indicate that
Intensive Interaction is ‘likely to be beneficial’ (Health Evidence Bulletins Wales, 2001,
p.12).
Present study
The study reported here took place in a service for adults where Intensive Interaction was
emerging practice (Samuel & Maggs, 1998; Samuel, 2001a & b). It addressed a gap in
the accumulating evidence on effectiveness by addressing two hypotheses:
1. Home-support staff as novice practitioners could learn to use the principles of
Intensive Interaction.
2. Intensive Interaction with novice practitioners would have a positive impact on a)
the communication and social abilities of people with profound intellectual
disabilities and b) the quality of the relationship between them.
From previous studies (particularly Kellett, 2001) a positive trend continuing across the
intervention phase as a consequence of improving skill was hypothesised, together with a
delay for behaviours emerging later in the usual developmental sequence.
5
Method
An interrupted time-series multiple-baseline across four participants design was used in
partial replication of Nind (1996) and Kellett (2001) with multiple measures used to aid
triangulation and minimise threats to validity. Whether novice practitioners could learn to
use the principles of Intensive Interaction was measured through video analysis of their
behaviours in interaction and analysis of the reflection records and responses to a staff
questionnaire. Impact on the communication and social abilities of the participants with
profound intellectual disabilities was measured through video analysis of their behaviours
in interaction combined with formal measures and analysis of qualitative staff data.
Impact on the quality of the relationship between novice practitioners and with people
with profound intellectual disabilities was tested through a related combination of direct
observation and indirect reporting.
A staggered baseline phase of six weeks commenced a week apart for each participant.
Questionnaires were administered at the start and end of the baseline phase [b1 and b6].
Following training an intervention phase of ‘formal’ Intensive Interaction (regular
sessions with post-session reflection) lasted twenty weeks with questionnaire probe at six
weeks [i6] and the end [i20]. Video data were collected at five points during the baseline
and twelve during the intervention. Session Reflection Records (Nind & Hewett, 2001)
and Historical Logs (of life events and changes in medication and health) were kept.
Settings & Participants
The settings comprised four bungalows in different towns with distinct home support
teams. The participants (see Table 1) had no previous involvement with Intensive
6
Interaction. Agreement for their participation was given by significant others. Staff
(Table 2) comprised three ‘practitioners’ and three ‘observers’ per participant. An
assistant psychologist visited weekly, filmed and functioned as an additional practitioner.
[Insert Table 1 & Table 2 about here]
Procedure
Baseline data collection comprised the participant usually first filmed ‘out of session’behaviour in the normal home environment when staff and co-tenants were not actively
interacting with the participant but available if she chose to initiate contact. After a fiveminute gap, a practitioner and assistant psychologist were each filmed engaging the
participant in ‘normal’ social interaction - not deliberately using principles or reflection
of Intensive Interaction.
Training comprised a half-day workshop facilitated by an expert practitioner as regularly
provided for the service together with provision of A Practical Guide to Intensive
Interaction (Nind & Hewett, 2001), Service Guidelines, Reflection Record Forms and
Support Group. Intervention comprised five sessions per week (duration unspecified, 100
in all). Sessions were followed by a brief reflective conversation with the assistant
psychologist, referring to recent Reflection Records.
Measures
Three practitioner behaviour codes were identified based on Clark and Seifer (1983) and
Nind et al. (2001): 'mirroring vocalisation and movement', 'contingent responding'; and
'forcing/overriding'. Participant behaviour codes were: 'visual scanning'; ‘looking at
7
face’; ‘engagement’ ('a state of ...arousal in the participant functionally related to the
behaviour of the practitioner' involving pleasure and expectancy 'regarding the potential
of the other person to cause affect') (Nind et al., 2001 p. 149); ‘joint focus; and ‘initiating
social/physical contact’. ObsWin (Martin et al., 1996) was used for analysis by the first
author and assistant psychologist, six undergraduate and four postgraduate psychologists
who were blind to the hypotheses and phase change. Percentage occurrence of behaviour
in each five-minute sample was calculated. A sample (12%) of video-coding was
subjected to inter- and intra-rater agreement checks. The Pre-Verbal Communication
Schedule (PVCS) (Kiernan & Reid, 1987) (Short Form) was completed at the four probepoints by an independent speech and language therapist with practitioners as informants.
The changing expectations of staff about Intensive Interaction were collected via postal
questionnaire before the study and at the four probe-points. Content analysis was
conducted by the first author, with a sample (33%) also coded by a volunteer research
assistant. The questionnaire incorporated two published criterion-referenced lists of (less
to more constructive) client engagement behaviours: the Interactive Sequence (Downs &
Craft, 1997): ‘resists’, ‘tolerates’, ‘cooperates passively’, ‘enjoys’, ‘responds
cooperatively’, ‘leads’, ‘imitates’ and ‘imitates independently’; and the Physical
Sociability Scale (Kellett, 2001, after Nind, 1996): ‘actively resists being held’, ‘resists
being held most but not all of the time’ through to ‘… initiates physical contact such as
clinging or grasping’.
Reflection Records data were analysed (by Norris, 2003) using the Interactive Sequence,
practitioner behaviour descriptors (Clark & Seifer, 1983) and coding into ‘neutral’,
‘negative’ or ‘positive’ for both the participant and practitioner.
8
Results
Fewer than 100 sessions were reported although practitioners indicated more occurred.
Duration varied as indicated in Table 3. [Insert Table 3 about here]
Inter- and intra-rater agreements are reported in Table 4. For the video data all mean
percentage agreements were at least 80%. Variable video quality and missing weeks
reduced the data. Trend analysis was impossible, but phase means were calculated and
interpreted with caution. The questionnaire return-rate ranged from 75% to 100%.
Historical Logs revealed ill health for all participants and caused late or missing filming.
[Insert Table 4 about here]
Hypothesis One
Video data indicate that practitioners learned to use mirroring of movements and
vocalisation and to use contingent responding more (see Table 5). The frequency and
extent of Reflection Record completion declined over time and their analysis showed
evidence of the use of principles of Intensive Interaction but not of progression.
[insert Table 5 about here]
Only half the practitioners attended a Support Group and all were reluctant to watch their
own videos, thus reducing the potential for learning through reflective practice. There
were indications of reflection, however with one observer commenting that 'the structure
has prompted staff to notice more and reflect'.
9
Hypothesis Two a)
Each participant developed differently (see Table 6). There was early evidence of impact
on looking behaviours, though for Alice this began during the baseline phase. The ability
to become engaged in social interaction and to do joint focus was more apparent later in
the intervention phase (see Table 7). Development of initiation of social/physical contact
was more patchy but as an event rather than a state and this is more difficult to reliably
capture.
[insert tables 6 & 7 around here]
The PVCS showed score improvements by the end of the intervention compared with
baseline for all participants for ‘positive interaction’ and improvements in ‘vocal
imitation’ by i6 for Alice and Diana. The end of the intervention also showed
improvement for Clare in ‘attention seeking’, ‘simple negation’ and ‘understanding of
non-vocal communication’. Other category scores varied inconsistently or were
unchanged.
The Interactive Sequence measure showed improvements by i6 for Betty, Clare (rated by
practitioners) and Diana (rated by observers). Improvement was noted at i20 for all four
participants. In the Physical Sociability Scale improved scores were evident by i6. Staff
Questionnaire data indicated an increasing expectation that Intensive Interaction would
enhance the skills of the participants and that it would gradually lead to success and
maintained progress. Reflection Records reported eye contact throughout for all
participants and, with the exception of Betty, frequent smiling. They noted some
developments before they were evident on video. Some comments affirmed progress,
10
such as Clare’s physiotherapist noting that she was more responsive and engaged in
interaction (i16), and some comments noted critical incidents, such as unusually (at i18)
Betty had reached out and touched a visitor’s hand as she lent on her chair when talking
to someone else.
Hypothesis Two b)
In Staff Questionnaire analysis the code applied most to practitioner data was 'team
cohesion' whereas 'benefits for staff (in general)' was applied most to observer data. The
code 'reciprocal relationship building (with participant)' fitted only three of 58
practitioner comments and three of 58 observer comments across the study. Concerning
changes in expectations about Intensive Interaction for the participants, across time,
practitioners made five of 68 comments coded as 'reciprocal relationship building',
whereas, the observers made no such comments. One practitioner at i20 commented ‘we
have learned to read each other’.
Reflection Record data generally rated the experience of Intensive Interaction as
‘positive’ for both participants and practitioners. At i16 it was reported that twice Betty
sought interaction with a practitioner 'moving herself through two rooms, climbing on to
the seat next to [her] and giving eye contact'. Previously Betty sat on sofas alone 'moving
…if anyone sits next to her…' and never sought physical proximity.
11
Discussion
Methodological Considerations
Methodological differences to earlier studies prevent direct comparison, for example, less
familiarity with filming and fewer datapoints than in Nind (1996) and procedural
differences in administration of the PVCS. These methodological differences alone could
account for the less marked progress. Less consistent positive change on the Physical
Sociability Scale could similarly reflect procedural differences, or practitioners being less
(consistently) facilitative of physical contact, or positioning themselves too far away for
participants with severe physical disabilities. The Interactive Sequence results did not
match the positive comments reported in the Reflection Records or anecdotally. Although
The Interactive Sequence had face validity, its reliability as a measure of change in
Intensive Interaction was not established.
The design of multiple-baseline across individuals in independent settings enables a
degree of replication that strengthens external validity. However, the baseline length was
pre-set and did not show stability thus weakening claims to intervention effect. Moreover,
the baseline instruction to interact ‘as they normally do’ in a context where staff did not
offer sustained social interaction with the participant was in itself an intervention,
perhaps explaining some of the baseline rises in ‘looking at face’ and ‘engagement’. As
in the previous studies interactive techniques were sometimes present during baseline,
hence the intervention phase was more clearly characterised by a change in intensity of
approach rather than a novel process.
12
Aspects of the intervention as well as the methodology differed from earlier studies.
Nind’s skilled practitioners were required to interact at least twice a day for a school year,
though Kellett noted that her novice practitioners' proposed daily sessions across the year
did not always occur. The present intervention lasted only twenty weeks with five
sessions requested per week; practitioners were inconsistently using Intensive Interaction
enabling processes even late in the intervention. Reflection was also a lesser part of the
intervention in this study than Nind's, though possibly not Kellett's; practitioners' analysis
of video data or even experiences was rare. This may have been better facilitated by more
skilled supervision than the assistant psychologist (herself a novice practitioner) could
offer.
Implications for clinical practice
The findings add to the evidence base, furthering knowledge about Intensive Interaction,
and also raise issues. Regardless of the ability of Intensive Interaction to make people
involved feel good, service demands that compete with sustaining and monitoring
Intensive Interaction may need to be addressed. Better planning and supervision, as
stressed by Irvine (2001) and Kellett (2001) may have made more impact in this study.
(Motivation and practice might be assisted by the new UK support staff qualification
(Carnaby, 2002), which promotes Intensive Interaction, and guidance now available on
video (Byers et al., 2002; Caldwell, 2002, 2005).) Record-keeping was sparse and more
specificity in recording formats may help to prompt practitioners to use principles they
overlook, though too task-orientated an approach may be counter-productive (Nind &
Hewett, 2005). Smaller cohesive teams ought to enable less variable practice, although
the questionnaire findings indicated idiosyncratic experience even within small teams.
13
Implications for research
Researching complex interventions presents real challenges as apparent in this study.
Replication with more attention to practitioner skill, a longer duration and across settings
might improve confidence in the results. Evaluating Intensive Interaction mirrors the
difficulties of multi-component psychotherapy process research in terms of variable
control. However in such research a sufficient level of practitioner expertise is assumed.
More sophisticated technology such as digital cameras, sensitive microphones and split
screens would enhance analysis and with more comprehensive data, sequential analysis
would be possible. Practitioners in this study rarely appeared to use traditional interactive
games such as 'peek-a-boo'. It is unclear if this was active reluctance because of
participant age, lack of expertise or for other reasons. More research is required into what
enhances and maintains practitioner interest and skill and what might be considered
minimum requirements for good enough practice. Longitudinal follow-up and evaluation
of any generalisation to other carers and to wider social inclusion would be timely.
Conclusion
The findings of this study complement and add to the existing evidence about the
development of communication and social ability for people with profound intellectual
disabilities through Intensive Interaction via partial replication in another context. Use of
Intensive Interaction in supported living by novice practitioners appears to offer some
potential, both for staff to learn to use some of the principles of the approach and for the
impact this might have on the communication and social abilities of the clients and their
relationship with them.
14
References
BYERS, R., DEE, L., HAYHOE, H. & MAUDSLEY, L. (2002) Enhancing Quality of Life.
Cambridge: University of Cambridge.
CALDWELL, P. (2002) Learning the Language. Brighton: Pavilion.
CALDWELL, P. (2005) with Stevens, P. Creative conversations. Brighton: Pavilion.
CARNABY, S. (2002) Your Pathway to Achievement: Trainer's manual for the learning
disability awards framework. Brighton: Pavilion.
CLARK, G. N. & SEIFER, R. (1983) 'Facilitating mother-infant communication: a treatment
model for high-risk and developmentally delayed infants', Infant Mental Health Journal
4(2): 67-82.
DOWNS, C. & CRAFT, A. (1997) Sex in Context: a personal and social development
programme for children and adults with profound a multiple impairments. Brighton:
Pavilion.
ELGIE, S. & MAGUIRE, N. (2001) 'Intensive Interaction with a woman with multiple and
profound disabilities: a case study', Tizard Learning Disability Review 6(3): 18-24.
EPHRAIM, G. W. (1986) A Brief Introduction to Augmented Mothering. Harperbury Hospital:
Playtrac.
HEALTH EVIDENCE BULLETINS: WALES February 2001. Retrieved 1 March 2007, from
http://hebw.cf.ac.uk/learningdisabilities/chapter2.htm
15
IRVINE, C. (2001) 'Interaction reaction: keeping Intensive Interaction going', PMLD Link 39:
13-15.
KELLETT, M. C. (2000) 'Sam's Story: the effect of Intensive Interaction on the social and
communicative ability of a young child with severe learning difficulties', Support for
Learning 15: 165-172.
KELLETT, M. C. (2001) Implementing Intensive Interaction. Unpublished PhD thesis, Oxford
Brookes University.
KELLETT, M. C. (2003) 'Jacob’s journey: developing sociability and communication in a young
boy with severe and complex learning difficulties using the Intensive Interaction teaching
approach', Journal of Research in Special Educational Needs 3(1).
KELLETT, M., & NIND, M. (2003) Implementing Intensive Interaction in Schools: Guidance for
practitioners, managers and coordinators. London: David Fulton.
KIERNAN, C. & REID, B. (1987) Pre-verbal Communication Schedule. Windsor: NFERNelson.
LEANING, B. & WATSON, W. (2006) 'From the inside looking out - an Intensive Interaction
group for people with profound and multiple learning disabilities', British Journal of
Learning Disabilities 34(2): 103-109.
LOVELL, D. M., JONES S.P. & EPHRAIM G. (1998) 'The effect of intensive interaction on the
sociability of a man with severe intellectual disabilities', International Journal of Practical
Approaches to Disability 22(2/3): 3-9.
MARTIN, N.T., OLIVER C. & HALL, S. (1996) ObsWin-observational data collection and
analysis for Windows& trade. Birmingham: University of Birmingham.
16
NIND, M. (1996) 'Efficacy of Intensive Interaction', European Journal of Special Needs
Education 11: 48-66.
NIND, M. & HEWETT, D. (1988) 'Interaction as curriculum', British Journal of Special
Education 15(2): 55-57.
NIND, M. & HEWETT, D. (2001) A Practical Guide to Intensive Interaction. Kidderminster:
British Institute of Learning Disabilities.
NIND, M. & HEWETT, D. (2005) (2nd edn) Access to Communication: Developing the basics of
communication in people with severe learning difficulties through Intensive Interaction.
London: David Fulton.
NIND, M., KELLETT M. & HOPKINS V. (2001) 'Teachers’ talk styles: communicating with
learners with severe and complex learning difficulties', Child Language Teaching &
Therapy 17(2): 145-159.
NORRIS, G. (2003) An evaluation of Intensive Interaction session reflection records. Small
Scale Research Project, Oxford Doctoral Course in Clinical Psychology. Unpublished
manuscript.
SAMUEL, J. (2001a) 'Intensive Interaction', Clinical Psychology Forum 148: 22-25.
SAMUEL, J. (2001b) 'Intensive Interaction in context', Tizard Learning Disability Review 6(3):
25-30.
SAMUEL, J. & MAGGS, J. (1998) 'Introducing Intensive Interaction for people with profound
learning disabilities living in small staffed houses in the community', in HEWETT, D. &
NIND, M., Interaction in Action, pp.119-148. London: David Fulton.
17
WATSON, J. (1994) 'Using Interaction in the education of pupils with PMLDs (ii) Intensive
Interaction: two case studies', in WARE, J., Educating Children with Profound and
Multiple Learning Difficulties, pp.150-159. London: David Fulton.
WATSON, J. & KNIGHT, C. (1991) 'An evaluation of Intensive Interaction teaching with pupils
with very severe learning difficulties', Child Language Teaching and Therapy 7(3): 310325.
18